Ghost DNA is genetic material inherited from ancient hominin populations that are known only through their footprint in modern genomes, not from any recovered bones or fossils. Unlike Neanderthals or the Altai Denisovan, whose ancient DNA has been directly sequenced from physical remains, ghost lineages are detected purely through statistical patterns in living people’s chromosomes. The concept has reshaped our understanding of human evolution, revealing that our ancestors interbred with more groups than the fossil record ever hinted at, and that some of those encounters left traces that still influence immunity, skin, and even how well certain populations tolerate thin mountain air.
Detecting Populations That Left No Bones Behind
The “ghost” in ghost DNA refers to invisibility in the fossil record. Researchers identify these lineages by analyzing stretches of DNA in modern people that don’t match any known archaic genome. When computational models try to explain the genetic variation in a living population using only known ancestors, they sometimes fail: segments of chromosomes look too old or too divergent to have come from modern humans, Neanderthals, or the single Denisovan individual whose genome has been sequenced. The mismatch points to a missing source population, one that contributed DNA through interbreeding but left no skeletal remains for archaeologists to find.
A 2020 study published in Science Advances demonstrated one approach: building genome-wide maps of archaic ancestry without needing a reference archaic genome at all. Instead, the researchers used patterns in how common or rare certain genetic variants are across a population, and compared those patterns to what models predict under different histories of interbreeding. When the data can’t be explained without adding an unknown contributor, that contributor becomes a ghost.1PubMed Central. Recovering signals of ghost archaic introgression in African populations The broader significance, as one review put it, is that ghost introgression appears across the tree of life and has “important consequences for the study of numerous evolutionary processes, including adaptation, speciation, and macroevolutionary patterns.”2PubMed. Ghost Introgression: Spooky Gene Flow in the Distant Past
Ghost Lineages in Africa
Much of the early conversation around archaic interbreeding centered on Neanderthal DNA in people of European and Asian descent. But some of the most striking ghost DNA findings have come from African populations, where interbreeding with known archaic groups like Neanderthals was minimal or absent.
The 2020 Science Advances study analyzed genomes from the Yoruba of West Africa and the Mende of Sierra Leone. It found that these populations derive roughly 2 to 19 percent of their genetic ancestry from an archaic population that split off from the lineage leading to both Neanderthals and modern humans. That divergence is ancient, possibly predating the Neanderthal-modern human split by hundreds of thousands of years.1PubMed Central. Recovering signals of ghost archaic introgression in African populations A separate analysis looking at a broader set of sub-Saharan populations estimated archaic introgression at about 4 percent in the Khoisan, about 4.3 percent in Mbuti Pygmies, and roughly 5.8 percent in the Mandenka, attributing it to an early-diverging and now extinct “ghost modern human lineage.”3PubMed Central. Whole-genome sequence analysis of a Pan African set of samples reveals archaic gene flow from an extinct basal population of modern humans into sub-Saharan populations
The difference between those two estimates isn’t necessarily a contradiction. The studies used different methods, different populations, and different models. One study’s “ghost” may represent a lineage that diverged before Neanderthals split off, while the other points to an early branch within the broader modern human family tree that went extinct. What they share is the conclusion that sub-Saharan Africans carry substantial DNA from populations no fossil has ever identified.
Multiple Denisovan Ghost Populations in Asia and Oceania
Denisovans are known from remarkably few physical remains, mostly fragments from a single cave in Siberia. Yet their genetic legacy in living people is vast and complicated. Research published in Nature Genetics found evidence of at least three separate introgression events from distinct Denisovan populations into modern humans, each showing a different degree of relatedness to the sequenced Altai Denisovan genome.4PubMed. A history of multiple Denisovan introgression events in modern humans In other words, the Denisovans weren’t a single homogeneous group. They included populations that had been separated from each other for long enough to become genetically distinct, and modern humans encountered and interbred with several of them independently.
This shows up most clearly in people from Near Oceania, including populations in Papua New Guinea and nearby islands. A 2025 study in Science reconstructed nearly 1.9 billion base pairs of archaic genome from these populations, including over 831 million base pairs of Denisovan sequence. The researchers confirmed introgression from three Denisovan-like groups in Near Oceanians.5PubMed Central. Long-term isolation and archaic introgression shape functional genetic variation in Near Oceania Some of these contributing populations are effectively ghosts: they are known only from the DNA they left in living people, not from any fossil. There is also evidence, from ancestral recombination graph analysis, that Oceanian populations carry deep lineages enriched in Denisovan regions that may reflect even older “super-archaic” gene flow, hinting at layers of interbreeding that go back further than previously appreciated.6PubMed Central. Recovering signatures of archaic introgression using ancestral recombination graphs
Breathing Thin Air Thanks to Denisovan DNA
One of the best-documented examples of archaic DNA actively helping modern humans involves a gene called EPAS1, sometimes called the “super athlete gene.” EPAS1 helps regulate how the body responds to low oxygen levels. In Tibetans, who have lived at extreme altitudes for thousands of years, a specific version of EPAS1 carries strong signatures of both positive natural selection and Denisovan introgression.7PubMed Central. The history and evolution of the Denisovan-EPAS1 haplotype in Tibetans The Tibetan version of the gene shares many variants with the Denisovan genome, even though the two lineages diverged roughly a million years ago.8PLOS Genetics. Evolutionary history of Tibetans inferred from whole-genome sequencing
What makes this case particularly clean is that, outside of EPAS1, the same study found no evidence that positive selection was acting on Denisovan-like segments elsewhere in Tibetan genomes.8PLOS Genetics. Evolutionary history of Tibetans inferred from whole-genome sequencing It wasn’t a broad wave of useful Denisovan DNA being retained. It was one gene, in one population, solving one specific environmental problem so effectively that natural selection drove it to high frequency while the rest of the Denisovan contribution faded. That kind of targeted benefit is what researchers mean by “adaptive introgression,” and it’s one of the clearest examples anywhere in human genetics of ancient interbreeding producing a tangible physiological advantage.
The same TRPS1 gene region found at high frequency in Near Oceanian populations also shows up in the African archaic introgression maps, where it reaches about 71 to 75 percent frequency in the Yoruba and Mende.9PubMed Central. Recovering signals of ghost archaic introgression in African populations – Section: DISCUSSION In the Oceanian context, TRPS1 is associated with skeletal development and appears to be under selection in populations as geographically scattered as central African rainforest hunter-gatherers and highland Ecuadorians.5PubMed Central. Long-term isolation and archaic introgression shape functional genetic variation in Near Oceania That convergence across continents and source populations is striking, and it suggests some archaic gene variants were independently useful in very different settings.
Skin, Hair, and UV Protection
Beyond high-altitude adaptation, some of the strongest evidence for beneficial archaic DNA involves skin and hair. A large Neanderthal-derived stretch of DNA on chromosome 9, encompassing a gene called BNC2, is found at about 70 percent frequency in Europeans. BNC2 is expressed in skin cells and has been linked to variation in skin pigmentation and freckling.10Current Biology. Ten years of Neanderthal introgression studies – Section: Skin and hair pigmentation A different Neanderthal-derived segment covering the POU2F3 gene on chromosome 11 shows up at about 60 percent frequency in East Asians but is nearly absent in Europeans. POU2F3 encodes a protein involved in the growth and specialization of skin cells. A third introgressed stretch on chromosome 3 includes the HYAL2 gene, which plays a role in how cells respond to ultraviolet radiation and contributes to skin pigmentation changes.10Current Biology. Ten years of Neanderthal introgression studies – Section: Skin and hair pigmentation
The pattern across these examples is instructive. Different populations retained different Neanderthal skin-related variants, likely because each variant was useful under the specific UV conditions those populations encountered as they spread across Eurasia. Neanderthals had already spent hundreds of thousands of years adapting to non-African environments. Modern humans arriving in those same environments could shortcut the process by picking up ready-made genetic tools through interbreeding.
Ghost DNA and the Immune System
Your immune system is arguably where archaic DNA has left its most complex mark, both helpful and harmful. Ancient interbreeding introduced gene variants that helped modern humans fight pathogens they had never encountered before. Research into the overlap between archaic introgression and immune function has revealed what’s been called a “poison-antidote” dynamic. When modern humans and Neanderthals came into contact, each group was exposed to viruses the other had been dealing with for millennia. Gene flow between the two populations allowed virus-interacting proteins that were already adapted to specific pathogens in the donor species to cross into the recipient, providing a measure of ready-made defense.11PubMed Central. Evidence that RNA viruses drove of adaptive introgression between Neanderthals and modern humans
This is an elegant model: interbreeding simultaneously created the problem (exposure to unfamiliar viruses) and delivered part of the solution (genetic resistance borrowed from the other species). But it also means the immune legacy of archaic DNA isn’t universally positive. Some inherited variants that were useful against ancient infections may behave differently in modern environments. Research into ancient genomics and immunity has identified a link between Neanderthal heritage and susceptibility to severe COVID-19.12PubMed Central. New insights into human immunity from ancient genomics A stretch of Neanderthal-derived DNA on chromosome 3 was associated with a heightened inflammatory response to SARS-CoV-2 infection. That same robust inflammatory tendency might have been highly effective against pathogens circulating tens of thousands of years ago, but it turned harmful in the context of a novel respiratory virus and modern intensive-care medicine.
Why Some Archaic DNA Sticks Around and Some Disappears
Not all ghost DNA has been treated equally by natural selection. Some archaic segments reached remarkably high frequencies, suggesting they conferred genuine advantages. In the Yoruba and Mende populations, the researchers identified dozens of loci where archaic segments appear in more than half the population, a frequency too high to explain by random genetic drift alone. Among them are segments near NF1, a tumor suppressor gene, present in roughly 83 to 85 percent of individuals. Other high-frequency archaic segments sit near genes involved in mitochondrial respiration in the testes, hormone regulation, and potassium channel function.9PubMed Central. Recovering signals of ghost archaic introgression in African populations – Section: DISCUSSION
On the other side of the ledger, large stretches of the genome have been swept clean of archaic DNA, sometimes called “deserts of introgression.” These are regions where any archaic contribution was presumably harmful and was rapidly purged by natural selection. In Eurasian populations, these deserts tend to cluster around genes critical for brain development and male fertility, areas where even small disruptions from a distantly related species could reduce an individual’s ability to reproduce. The picture that emerges is one of editing: interbreeding introduced a broad sweep of archaic variants, and then tens of thousands of years of selection kept the useful ones and discarded the rest.
How Ghost DNA Complicates the Tree of Human Evolution
The traditional model of human evolution, even the more nuanced recent versions, tends to depict it as a branching tree: lineages split, some go extinct, and one leads to us. Ghost DNA reveals that the real picture looks more like a tangled web. Lineages split and then reconnected through interbreeding, sometimes repeatedly, sometimes with partners whose very existence would be invisible without genomic analysis.
In Africa, the ghost lineages detected so far appear to have diverged from our lineage at different times and in different ways. One may predate the Neanderthal-modern human split entirely. Another may represent an early branch within the broader modern human family that later went extinct. In Southeast Asia and Oceania, at least three genetically distinguishable Denisovan-like populations contributed DNA to living humans, despite there being virtually no Denisovan fossil record outside a single Siberian cave and a jawbone from the Tibetan Plateau.4PubMed. A history of multiple Denisovan introgression events in modern humans In each case, the genomic evidence came first. The fossils, if they exist at all, haven’t been found yet.
This has practical implications for how scientists reconstruct evolutionary history. Methods that assume a strictly branching tree can miss or misinterpret signals of interbreeding, producing misleading estimates of when populations diverged or how large they were. Newer computational approaches, like those using ancestral recombination graphs, are designed to handle the kind of tangled histories that ghost introgression implies.6PubMed Central. Recovering signatures of archaic introgression using ancestral recombination graphs
Ethical Dimensions of Studying Ghost Ancestry
The populations most affected by ghost DNA discoveries are often Indigenous communities in Africa, Oceania, and the Americas, groups whose histories have already been distorted or erased by colonialism. When researchers announce that a population carries a percentage of DNA from an unknown archaic source, the framing matters enormously. A finding about “archaic ancestry” can be misrepresented to imply that some living people are somehow less modern or more “primitive” than others, which is both scientifically wrong and socially dangerous. Every living human is equally modern; archaic introgression is part of all major populations studied so far.
Recent work in archaeogenetics has emphasized the potential of ancient DNA research to uncover past histories in regions where colonialism has neglected or suppressed Indigenous oral histories.13Science. Going local with ancient DNA: A review of human histories from regional perspectives But that potential is only realized when research is conducted in genuine partnership with the communities whose genomes are being studied. Some of the most productive collaborations have involved Indigenous researchers and community advisory boards from the start, ensuring that findings are communicated in ways the community can use and that the narrative doesn’t get hijacked for purposes the participants never intended.
Ghost DNA Beyond Humans
Ghost introgression isn’t unique to our species. The same statistical tools that uncovered ghost lineages in human genomes are being applied across the tree of life, and they keep finding tangled webs of ancient gene flow. The concept extends beyond introgression from other species, too. In paleontology, researchers have used the idea of “ghost ranges” to describe the detectable presence of species that appear absent in the macrofossil record. Environmental DNA recovered from permafrost sediments, for example, showed that woolly mammoths and horses persisted in interior Alaska until at least 10,500 years ago, thousands of years later than their last known fossils in the region.14Proceedings of the National Academy of Sciences. Ancient DNA reveals late survival of mammoth and horse in interior Alaska
The parallel is worth noting because it captures the core idea behind all ghost DNA research: absence of physical evidence is not evidence of absence. In genetics, as in paleontology, organisms and lineages can be invisible to conventional methods yet detectable through molecular traces. As genome sequencing becomes cheaper and analytical methods more sophisticated, the list of ghost populations contributing to living species, human and otherwise, is almost certain to grow longer. The tree of life, it turns out, has always been more of a thicket.